Clinically we have attempted to develop novel agents that alter the biology of the cancer. Specifically, we have been interested for many years in inhibiting angiogenesis as a means to treat prostate cancer. The progression of prostate cancer from a latent to an aggressive form depends on the acquisition of the angiogenic phenotype. Without angiogenesis, the primary prostate tumor is confined to 1-2 mm in size, and remains indolent. Angiogenesis is also required at sites of secondary colonizations in order for prostate cancer metastases to proliferate and expand. Prostate tumors found in autopsy specimens from men without clinical prostate cancer have very low blood vessel content, compared to prostate cancer specimens from men with clinically-evident disease. Siegal and colleagues reported that microvessel density (MVD) was significantly higher in prostate cancer tissue, than in adjacent hyperplastic or benign tissue. Numerous studies have been conducted to evaluate the use of MVD in prostate cancer samples as a prognostic and/or diagnostic marker. Most studies have demonstrated that MVD does help predict pathological stage and patient outcome. Using specimens from radical prostatectomies, Weidner et al., correlated increased angiogenesis in primary tumor specimens with metastatic disease. Several other studies also found MVD to independently predict the outcome of patients with prostate cancer. A recent study did not find MVD to be a useful prognostic indicator for men with clinically localized prostate cancer. Angiogenesis is driven by an imbalance of positive and negative regulators. One of the most potent positive regulators of angiogenesis is the vascular endothelial growth factor (VEGF). Prostate cancer cells produce VEGF at very high concentrations compared to normal prostate tissue. Such elevated levels of VEGF contribute to prostate cancer progression by inducing angiogenesis in the stroma via paracrine signalling. Using different sublines of LNCap cell lines, Balbay and colleagues demonstrated that the metastatic potential of human prostate cancer cell lines in an athymic mice model correlated with their VEGF expression. VEGF production is regulated by androgens in both normal and malignant prostate tissues. When prostate cancer cells progress to an androgen-independent state, VEGF regulation by androgens is also lost. Cellular hypoxia then becomes the main regulator of VEGF. VEGF acts upon two high affinity tyrosine kinase family receptors, Flt-1/ VEGFR-1 and Flk-1/ VEGFR-2. While previously believed to be specific to endothelial cells, these VEGF receptors have recently been localized to several types of tumors, including prostate. A recent study reports that Flt-1 is present in BPH and PIN, but lost in prostate cancer cells and with tumor dedifferentiation, implicating a role for this receptor in prostatic transformation to malignancy. Another member of the VEGF family, VEGF C, which binds to VEGF receptor-3 (VEGFR-3/ Flt-4) is also produced by prostate cancer cells and has recently been implicated in lymph node metastasis. Thus, the strong interplay between prostate cancer progression and angiogenesis are quickly being realized as this field unfolds. Antiangiogenic agents which we have clinically evaluated include: suramin, CAI, thalidomide, TNP-470, COL3, and somatuline. Currently, we are assessing docetaxel with or without thalidomide, and ketoconazole with or without alendronate (an MMP inhibitor) in patients with androgen indepedent prostate cancer. We have initiated a phase I clinical trial with CC5013 and 2ME (angiogenesis inhibitors). The angiogenic property of thalidomide reported by D'Amato and colleagues has prompted its clinical evaluation in various solid tumors including prostate cancer. Our laboratory previously showed that one of the products of cytochrome P450 2C19 isozyme biotransformation of thalidomide, 5'-OH-thalidomide, is responsible for the drug's antiangiogenic activity. Based on the chemical structure of this metabolite, we have synthesized 118 analogs of thalidomide and have evaluated them using four in vitro models to assess activity in the inhibition of angiogenesis (rat aorta model, human saphenous vein model, cultured endothelial cells, and tube formationassay). We have identified the most potent of these and have patented them. We are continuing to develop these compounds. These compounds appear have minimal side effects in initial preclinical toxicology studies.We chose to combine thalidomide with docetaxel. Using a randomized Phase II trial design we compared weekly docetaxel (30 mg/m2) with or without 200 mg/d of thalidomide. The objective of this study was to determine whether the combination of thalidomide and docetaxel could produce a sufficiently high clinical response rate to warrant further investigation. All of the patients had metastatic AIPC and were progressing on androgen blockade, as well as anti-androgen withdrawal. A total of 75 patients were enrolled onto this trial, 25 patients in docetaxel alone arm and 50 patients in the combination arm. A two-stage mini-max design, that is one in which the design sought to minimize the maximum number of patients enrolled, was chosen for both arms. We used P0=0.25 (undesirable response rate) and P1=0.45 (target response rate) for the combination arm, and P0=0.05 and P1=0.025 for the docetaxel-alone arm, with ?=0.05 and ?=0.10 for both arms. We used the PSA working group consensus criteria combined with radiographic studies to determine the proportion of patients with a PSA decline and time to progression. Both at the midpoint evaluation and at the conclusion of the trial, the proportion of patients with a >50% decline in PSA was higher in the combination arm (25 of 47 patients, 53%) than in the docetaxel alone arm (9 of 24 patients, 37%). These response rates satisfied criteria for further evaluation (combination arm) and for consistency with prior results (single agent arm). Likewise, 3 of 11 patients (27%) in the docetaxel-alone arm with measurable soft tissue disease by CT scan developed a partial response, and 7 of 20 patients (35%) in the combined treatment group had a partial response in soft tissue disease. None of the patients with bony lesions had a normalization of their bone scan. The median progression free survival was 3.7 mo and 5.9 mo in the docetaxel and combined group, respectively (p=0.32 for the overall difference in the curve). The 18 mo survival was 42.9% in the docetaxel alone group and 68.2% in the combined group. The median overall survival in the docetaxel alone group was 14 mo compared with 28 mo for the combination arm (p=0.11)Based on the positive data we found with the combination of thalidomide and docetaxel, we have recently initiated a Phase II study of estramustine, docetaxel and thalidomide in chemo-nave patients with AIPC. The primary objective of the study is to show a statistically significant improvement in PSA decline when compared to the combination of docetaxel and thalidomide. We are also looking at multiple secondary endpoints. These include possible pharmacokinetic interactions among the study agents, and potential correlation between patient genotype (CYP2C19) and efficacy of treatment. We are also looking for circulating tumor cells in blood before and after treatment. Additionally, we will be monitoring the tolerability of the regimen and survival duration as endpoints. Thus far, we have enrolled 7 patients, all of whom have exhibited a PSA reduction of >50%. The regimen has been well tolerated, although there have been several dose reductions of thalidomide and/or estramustine due to side effects (e.g., fatigue, gastrointestinal (GI) complications).Dr. Dahut and myself have recently initiated a Phase II trial of thalidomide, docetaxel, prednisone and bevacizumab in chemo-nave AIPC patients.